Trust during AI rollouts: How HR keeps pace without losing people

It’s only natural to fear what we don’t fully understand. But when it’s coupled with the threat of job loss and the sense that what you do no longer matters, it threatens the very core of who your employees are as humans.

You only need to look at recent headlines to see why employees are feeling this fear so acutely. Multiple organisations have now set out their vision for an AI-enabled future. Some have made mass layoffs into the thousands — others have asked managers to prove the value of each new human hire against an AI counterpart.

And beneath all of that, employees feel like the burden of proof is on them to show why their role and skills still matter.

These fears tend to pop up as four key questions:

• “Will I still have a job in a year?” AI is automating tasks across every function. When organisations avoid honest conversations about what that means for specific roles, it feeds the fear that employees will lose their livelihoods.

• “How will my job change?” Even employees that feel secure in their roles can feel unsettled by not knowing what their day-to-day will look like in the near future. When roles aren’t redefined clearly, people fill the gaps with worst-case assumptions.

• “Am I going to get left behind?” The fear of being left behind is very real — an April 2026 article in The Independent reported that older workers are retiring early rather than learning AI-native skills. Without clear direction on reskilling or development, AI adoption can feel like a threat rather than a motivator.

• “Has anyone asked how I’m feeling?” When changes are handed down at a leadership level and employees have no input, employees feel that change is happening to them, rather than bringing them along.

When these questions go unanswered, employees create their own stories on what’s actually happening. The result is a direct hit on the organisation’s bottom line. 

Gallup’s State of the Global Workplace 2026 report found that employee engagement has hit its lowest point since 2020. Personio's 2025 Workforce Pulse research found that employees worried about losing their job to AI are 39% more likely to be job hunting. Those uneasy about AI-driven workforce changes are 29% more likely to look elsewhere. 

And the cost of losing people adds up. According to Personio research, European companies lost €274k (£240k) in 2025 due to preventable people problems — including turnover, skills gaps, and workforce planning.

When scaling Facebook in the company’s early days, Mark Zuckerberg coined a phrase that became a shorthand for scaling businesses everywhere: move fast, break things. It described the trade-offs and choices most businesses had to make to see success.

When we’re talking about AI, we often see that narrative disguised as speed versus trust — choose one at the cost of the other.

It’s a false dichotomy. Because the organisations pulling ahead in the race to roll out AI aren’t the ones going faster at all costs — they’re the ones making trust central to how they operate.

“Speed and trust aren’t in opposition to each other. It really comes back to a question of where are you today — are you coming from a high trust or low trust base? That ultimately will determine whether your speed and trust are in opposition. They’re not in opposition to each other by nature.“

Lenke Taylor

Chief People Officer, Personio

That baseline matters even more during an AI rollout. AI-driven change doesn’t behave like a typical change management programme. There’s no defined scope, clear destination, or a point at which change stops. Because of the speed at which the technology is changing, organisations are now locked into a continuous loop where workflows, roles, and capabilities keep evolving. And when change is the new constant, moving fast becomes less important than building the momentum to keep adapting. 

This is why during AI-driven transformation, forward-thinking HR leaders are designing for the and, not the or — sustaining the trust that sustains momentum.

HR may not be responsible for setting the pace of AI adoption, but it can shape how the organisation handles the human side of change. This hinges on creating the processes, conditions, and behaviours for trust to stay high while change is the new normal.

AI-driven transformation is a whole-organisation change management process. But your organisation is made up of people, not computers. This makes bringing HR in early one of the most impactful shifts an organisation can make when driving a long-term return on investment with AI. 

But one of the most pervasive problems is that HR often isn’t included in the AI rollout memo. And worse, HR’s own hesitations around AI could be holding them back from being able to act effectively.

“The HR function is accountable for the most human dimensions of how people work. But as I was talking with people leaders, there were a lot of stakeholders telling me ‘I'm not heavily involved in our AI agenda. I'm trying to stay clear of it. I'm either not comfortable with AI or I'm not being included in those conversations.’“

Jess Larsen

CPO and Founder, Thriving Humans

When HR is involved in AI rollouts early, teams can flag legal considerations, anticipate how it will impact the employee experience and future workforce needs, and manage the emotional impact. Positioning HR as the arbiter of workforce risk is key — because it frames the consequences of supporting your people through change.

Pro tip: Don’t wait to be invited. Gather information from technical leads, department heads, and people leaders on what’s happening on the ground to understand how the rollout is impacting your people. 

Clarity and certainty tend to get confused in AI rollouts. Employees rarely expect certainty during change. But they do need clarity on what’s changing, what isn’t, and what you’re still figuring out.

This approach sets expectations without overpromising. But the important thing is in how leaders tap into the emotional undercurrent of their organisation. It’s a way of normalising vulnerability and learning in public — saying, "hey, I’m not sure about this either".

“It’s really hard for businesses to declare ‘this is the roadmap, and this is what you can expect’. It’s much more of a case of, ‘this is the direction we’re going in, these are the principles by which we’re going to work, and this is how we’re going to learn as we go’. Having leaders share ‘These are the things I’m struggling with' or 'We made a mistake’ helps people understand perfection isn’t the goal — the goal is learning. That can do wonders for shifting your culture.“

Jess Larsen

CPO and Founder, Thriving Humans

Pro tip: Build failure into your communications. Share wins and failures across company-wide channels, including Slack and all hands. Invite employees to talk through experiments and projects, share what they tried, what didn’t work, and what they learned.

Alongside clarity on what’s happening at an organisational level, employees need clear expectations on how to use the tools you’re rolling out. 

Governance and ethics, including concrete examples of what is and isn’t allowed, are essential for protecting proprietary data and trade secrets. But to truly bring your employees along, you need to make them owners.

This applies to output as well as input. At Personio, employees are ultimately responsible for assessing the truth of what the AI generates, and how they use it. This turns employees into active participants, not passengers.

“One of the priorities for us was helping people understand that the purpose of AI is to start with your idea: How do I actually start with my own voice and use the AI to help me clarify or develop that idea further?“

Lenke Taylor

Chief People Officer, Personio

Pro tip: Set expectations early on how employees should work with AI-generated content in the tools — for example, performance feedback or development plans. Make sure employees understand their role in starting from a human-first perspective and sense-checking all AI outputs.

Uncertainty and fear are often contagious. And in a time when change is constant, your people data is one of your earliest indicators as to how it’s landing with your workforce.

Regular pulse surveys and manager one-on-ones can help HR track employee sentiment and spot when uncertainty turns into disengagement and distrust before it hits retention. But viewed alone, this can only show you what’s already happened — and it only works when employees feel safe giving feedback in the first place. 

The key is pairing that information with the trust signals that your people analytics can’t capture — how people behave in day-to-day life at your company. That could be as much about what your employees say or do as what they don’t.

“Look at employee behaviours when leaders speak. Do people believe what they say automatically? Is there a genuine agreement, or is there a conversation out in the hall that's different? How comfortable are people to openly challenge? Not getting a reaction is actually often a very negative indicator — either people don't feel comfortable saying stuff, they don't think it's going to be listened to, or they've just checked out completely.“

Jess Larsen

Founder and CPO, Thriving Humans

Pro tip: Stay and exit interviews can be a rich source of unfiltered insight into what hasn’t worked with your AI rollout and strategy. Set aside some time in each to talk in general terms about how employees feel about their day-to-day work and role.

Want to hear about how other HR leaders are navigating the AI shift? Watch our on-demand event: Moving at the speed of AI without breaking your people.

Most organisations still see AI as the current race they need to win — faster adoption, faster returns, faster everything. As the tech only continues to increase in sophistication, this pressure isn’t going away. But your people might if you don’t bring them with you.

Trust isn’t something you start building during an AI rollout — and organisations with stronger foundations will naturally find it easier to gain momentum. But trust isn’t binary, and the decisions you make during your rollout can strengthen what’s already there when certainty is in short supply. Prioritising clarity, employee agency, and visible follow-through on the unanswered questions will help maintain both trust and momentum for change.

That work starts from where your organisation is now.

“Nobody is up to speed. It's a great time to start wherever you are. “

Jess Larsen

CPO and Founder, Thriving Humans

Get the frameworks, data, and expert insights you need to lead AI transformation with confidence.

Watch our on-demand AI event to learn more.

Start with the practical things first. Outline what tools you’re rolling out, usage expectations, and what will change day to day. Explain why the organisation is investing in AI. Then, communicate what you’re still figuring out — whether that’s how roles will change, how you’ll upskill employees, or how internal processes will change.

Avoid silence, but don’t over-reassure, either. Acknowledge what people are worried about out loud, and be clear about what you do and don’t know. Focus on normalising the worries and framing the rollout in terms of your organisation’s direction and principles, rather than allowing employees to dwell in fear.

Managers help employees navigate the day-to-day reality of an AI rollout. That includes managing workflows and workloads, and helping employees understand how roles and tasks may change at a team level. But they also provide critical emotional support — listening to employee concerns during one-on-ones, creating space for questions, and escalating concerns up to leadership as required. 

Track a combination of data points across the employee lifecycle. Engagement scores and survey feedback, particularly at a team or manager level, will give you both qualitative and quantitative insights on how employees are feeling. Behavioural signals, like tool adoption rates and training participation, will give you a sense of uptake. Keep an eye on lagging indicators, like performance and attrition, to see if patterns emerge over time.

Regularly enough to track patterns in engagement and overall AI confidence — but not so often employees get survey fatigue. Run a quarterly AI-specific pulse survey to help you understand employee confidence and adoption rates. Pair that with an additional question in your regular engagement survey to surface qualitative feedback and concerns as they happen.